126 



the ripe fruit had fallen. The ants had converted each peduncle 

 into a nest by providing it with a small circular opening, and 

 removing the pulpy tissue from its interior. In all these nests, 

 and attached to their walls, were several small reddish Coccids, 

 the excrement of which is probably an important part of the 

 ant's food. I did not see the Solenopsis out on the stems and 

 foHage, but even if they are in the habit of frequenting the sur- 

 faces of the plant, they are probably completely overlooked by 

 the PseudomyrmcB on account of their minute size. Similar 

 small yellow species of Solenopsis (5. fugax Latr., 5. molesta Say, 

 etc.) are known to live in the walls of the earthen nests of various 

 European and North American ants and to prey on their larvae 

 and pupae. It would be interesting to know whether the acacia 

 Solenopsis ever assumes a similar lestobiotic relation towards Ps. 

 fidvescens. 



I believe that we may regard the relationship existing be- 

 tween Ps. fidvescens and gracilis on the one hand, and Camponotus 

 planatus on the other, as one of parabiosis. This term was first 

 introduced by Forel in i8g8 to designate a peculiar pacific 

 relationship between Dolichoderus dehilis and Crematogaster 

 parahiotica. He found these two ants in Colombia nesting 

 in an abandoned termite nest, in such a manner that each species 

 kept its brood together in its own chambers and galleries, but 

 the chambers and galleries of the one species interdigitated and 

 even inosculated with those of the other in a very intimate man- 

 ner. The two species, moreover, foraged together on the same 

 plants, either separately or in a common file. Each of them was 

 also found nesting by itself. I may say in passing that I found 

 no less than a dozen colonies of these same ants in Panama and 

 Guatemala, and in all cases their nests presented essentially the 

 same pecuHarities as those described by Forel. Their foraging 

 habits, too, conformed with his description. Mann (1912) has 

 very recently described similar parabiotic relations between 

 certain Brazilian ants [Dolichoderus bispinosus Oliv, and Cre- 

 matogaster sp. ; Odontomachus affinis may i Mann and Dolicho- 

 derus dehilis var. rufescens Mann). Some years ago I also 

 included under the head of parabiosis the cases in which different 

 species of neotropical ants show a tendency to inhabit the same 

 Tillandsias and other BromeUaceous epiphytes. To these various 



